US9633269B2 - Image-based liveness detection for ultrasonic fingerprints - Google Patents

Image-based liveness detection for ultrasonic fingerprints Download PDF

Info

Publication number
US9633269B2
US9633269B2 US14/845,174 US201514845174A US9633269B2 US 9633269 B2 US9633269 B2 US 9633269B2 US 201514845174 A US201514845174 A US 201514845174A US 9633269 B2 US9633269 B2 US 9633269B2
Authority
US
United States
Prior art keywords
ridge
valley
image
feature vectors
feature
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
US14/845,174
Other languages
English (en)
Other versions
US20160070968A1 (en
Inventor
Jin Gu
Ming Yu Chen
Eliza Yingzi Du
Kwokleung Chan
Esra Vural
Saurav Bandyopadhyay
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qualcomm Inc
Original Assignee
Qualcomm Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US14/845,174 priority Critical patent/US9633269B2/en
Application filed by Qualcomm Inc filed Critical Qualcomm Inc
Priority to PCT/US2015/048583 priority patent/WO2016037078A1/en
Priority to CN201580046844.2A priority patent/CN106796650B/zh
Priority to CN201580045104.7A priority patent/CN106663198A/zh
Priority to KR1020177005953A priority patent/KR102467916B1/ko
Priority to PCT/US2015/048581 priority patent/WO2016037077A1/en
Priority to EP15772060.8A priority patent/EP3189470A1/en
Priority to JP2017511746A priority patent/JP6656228B2/ja
Priority to BR112017004379-3A priority patent/BR112017004379B1/pt
Priority to AU2015311785A priority patent/AU2015311785B9/en
Priority to CA2957218A priority patent/CA2957218C/en
Priority to EP15772061.6A priority patent/EP3189471B1/en
Priority to KR1020227039636A priority patent/KR102636547B1/ko
Assigned to QUALCOMM INCORPORATED reassignment QUALCOMM INCORPORATED ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: VURAL, Esra, CHAN, KWOKLEUNG, DU, ELIZA YINGZI, GU, JIN, BANDYOPADHYAY, SAURAV, CHEN, MING YU
Publication of US20160070968A1 publication Critical patent/US20160070968A1/en
Application granted granted Critical
Publication of US9633269B2 publication Critical patent/US9633269B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G06K9/00906—
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/40—Spoof detection, e.g. liveness detection
    • G06V40/45—Detection of the body part being alive
    • G06K9/0002—
    • G06K9/00093—
    • G06K9/001—
    • G06K9/00107—
    • G06K9/00114—
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/32—Determination of transform parameters for the alignment of images, i.e. image registration using correlation-based methods
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T7/41—Analysis of texture based on statistical description of texture
    • G06T7/44—Analysis of texture based on statistical description of texture using image operators, e.g. filters, edge density metrics or local histograms
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/13—Sensors therefor
    • G06V40/1306—Sensors therefor non-optical, e.g. ultrasonic or capacitive sensing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1365—Matching; Classification
    • G06V40/1371—Matching features related to minutiae or pores
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1365—Matching; Classification
    • G06V40/1376—Matching features related to ridge properties or fingerprint texture
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
    • G06V40/1388—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger using image processing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10132—Ultrasound image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20064—Wavelet transform [DWT]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20072—Graph-based image processing

Definitions

  • the present disclosure relates to devices and methods for detecting whether a biometric has been provided by a live-being.
  • Biometric sensors are widely used to distinguish between individuals that are authorized to engage in an activity and individuals that are not authorized to engage in that activity. For example, fingerprint sensors are commonly used to determine whether a fingerprint provided by an individual matches information in a database, and if a match is determined, then the individual may be allowed to engage in an activity. For example, the individual may be allowed to enter a building or room, or allowed to use an electronic device such as a mobile phone or an application running on a mobile device.
  • Biometric sensors can be deceived (a.k.a. “spoofed”), and thereby permit an unauthorized individual to engage in an activity that is reserved for authorized individuals.
  • Spoofing a fingerprint sensor may be accomplished in different ways. These include using a fake fingerprint, using body parts other than a finger, and using a dead finger from a person. Since it is unlikely that the particular type of spoofing to be used on a fingerprint sensor will be known in advance, it is important to guard against all types of spoofs.
  • One method of guarding against spoofing involves determining whether the biometric object exhibits characteristics associated with liveness.
  • Liveness detection is important because many methods of deceiving an identification system and/or verification system use spoofs that are not alive. For example, a latex finger may be made to have ridges and valleys resembling a fingerprint of an authorized user. When such a latex spoof is presented to a fingerprint sensor, the scanning system may falsely conclude that the latex spoof is the finger of an authorized user, even though the latex spoof is not part of a living being. Therefore, a system that is able to detect whether a biometric object is part of a live being would be useful in detecting the presence of a spoof.
  • the individual presenting the biometric object may be denied access (for example, access to a requested activity) even if the biometric information (such as a fingerprint) matches information in an associated identification/verification database.
  • One implementation of the present disclosure is a method of detecting liveness.
  • the method may comprise obtaining a single ultrasonic image of a biometric object.
  • the method may further comprise subdividing the single ultrasonic image into a plurality of overlapping sample blocks.
  • at least a subset of the plurality of sample blocks overlap with at least another of the subset of the plurality of sample blocks.
  • the method may further comprise extracting feature vectors in a spatial domain and a frequency domain from each of the plurality of sample blocks.
  • the feature vectors may include a histogram feature.
  • the histogram feature may include ultrasonic impedance characteristics of ridges and valleys of the biometric object.
  • the histogram feature may also include one or more of the following: ridge energy, ridge entropy, ridge variance, ridge skewness, ridge kurtosis, ridge variance coefficient, valley energy, valley entropy, valley variance, valley skewness valley kurtosis, and valley variance coefficient.
  • the method may further comprise comparing the feature vectors from each of the plurality of sample blocks to a classification model.
  • the feature vectors include an image-quality measurement.
  • the image-quality measurement may include one or more of the following: ridge orientations smoothness, ridge orientation certainty level, and a Gabor-based feature.
  • Comparing the feature vectors may include weighting a first set of the plurality of sample blocks relative to a second set of the plurality of sample blocks.
  • the feature vectors may also include cross-ridge-valley features.
  • the cross-ridge-valley features may include one or more of the following: ridge and valley separation in gray scale values, ridge and valley width, ridge-to-valley transition rate, valley-to-ridge transition rate, a frequency spectrum of a Fourier transform of a gray scale profile along a direction, and a ratio of image brightness of ridges of the biometric object to image brightness of valleys of the biometric object.
  • the feature vectors may also include an image-pattern feature, for example, a wavelet-based feature.
  • the feature vectors may include ratio-based features.
  • the ratio-based features may include a ratio of individual features between ridge and valley of the biometric object.
  • the ratio-based features may include one or more of the following: ratio of mean of ridge and valley, ratio of median of ridge and valley, ratio of energy of ridge and valley, ratio of entropy of ridge and valley, ratio of variance of ridge and valley, ratio of skewness of ridge and valley, ratio of kurtosis of ridge and valley, and ratio of variance coefficient of ridge and valley.
  • the method may further comprise altering the classification model based upon the comparison of the feature vectors to the classification model.
  • the method may further comprise altering the classification model based upon the single ultrasonic image of the biometric object.
  • the disclosure may also be embodied as a method or system for determining whether a biometric was obtained from a live individual, and thus may be used to determine whether the biometric permits an activity.
  • a method or system may employ vectors associated with liveness data based on histogram features, ratio-based features, cross-ridge-valley features, imaging quality features, and/or special image patterns, such as wavelet-based features.
  • Textural features such as histogram-based statistics, may be calculated for both ridges and valleys present in the biometric.
  • a block-based method of classification may be used.
  • feature vectors such as those computed from one or more of the five groups identified above
  • a decision regarding liveness is made with respect to each of the sample blocks.
  • a decision regarding whether a particular image came from a live individual or not may be made by fusing (e.g., mathematically combining based on a determined algorithm) the classification decisions from each of the sample blocks.
  • a subject-specific classification method In such a method, a set of features derived from images known to come from a particular individual are obtained and identified as being good liveness indicators for that individual.
  • the set of features may be derived from scan data, or collected non-image data.
  • Machine learning algorithms could be used to build the subject-specific classifier.
  • a high level learning methodology to identify those features that are good liveness indicators and build the classifier, it may be possible to reduce the computation of feature extraction by specifying the extraction of only those features that have been determined to be good liveness indicators for that individual.
  • Advantages of the method or system may include one or more of the following: (a) single-image based method for liveness detection; (b) resistance to the effects of noisy images or problematic fingerprints; and (c) liveness determinations that do not require perfect segmentation of ridges and valleys because the system or method does not depend on an accurate ridge/valley mask.
  • FIG. 1 is a schematic depicting how a liveness detection system/method may be part of a larger system/method
  • FIG. 2 is a flowchart illustrating that methodologies for determining liveness may be part of an enrollment process, as well as a matching/recognition process;
  • FIG. 3 is a flowchart showing additional detail about how methodologies for determining liveness might be utilized in a matching/recognition process
  • FIG. 4 is a flowchart showing additional detail about how methodologies for determining liveness might be utilized in a matching/recognition process
  • FIG. 5A depicts a collected fingerprint image
  • FIG. 5B depicts a ridge-valley mask
  • FIG. 6A depicts a fingerprint image with a vector superimposed on the image
  • FIG. 6B depicts a brightness profile taken along the superimposed vector in FIG. 5A ;
  • FIG. 7 is a flowchart depicting a baseline method for liveness detection
  • FIG. 8 is a flowchart depicting a liveness enrollment method for liveness detection
  • FIG. 9 is a flowchart depicting operation of a liveness detection method
  • FIG. 10 depicts an image block and superimposed thereon are multiple sample blocks
  • FIG. 11 is a flowchart illustrating a method of sample block division and classification model creation/updates
  • FIG. 12 is a flowchart illustrating a method of updating a classification model
  • FIG. 13 is a flowchart illustrating one implementation of a method for determining liveness.
  • the application may be embodied as methods and systems for detecting liveness.
  • liveness can be detected based upon the use of a plurality of image-based features that are derived from a biometric, such as a fingerprint.
  • the image-based features may be in the spatial and/or frequency domains, and may be used to determine whether a particular biometric is a real or a fake biometric.
  • a biometric such as a fingerprint
  • the image-based features may be in the spatial and/or frequency domains, and may be used to determine whether a particular biometric is a real or a fake biometric.
  • certain aspects of the application will be described with reference to fingerprints and ultrasonic fingerprint sensors, but the application is not limited to these.
  • FIG. 1 is a flow diagram depicting a method that includes determining liveness.
  • Image information about a biometric object may be acquired 100 and analyzed 103 in order to determine whether the biometric object is part of a live individual.
  • the image information used to determine liveness may be the same information used to generate an image of the object.
  • the image information may be information about a fingerprint that was acquired 100 using an ultrasonic sensor, such as described in U.S. Pat. No. 7,739,912 entitled “Ultrasonic Fingerprint Scanning Utilizing a Plane Wave.”
  • Analysis of the image information may be carried out by one or more multipurpose computers or processors that have been programmed to analyze the image information, such as one or more applications processors of a mobile device.
  • the acquired image information or data may be pre-processed as needed, for example, to reduce noise, increase contrast, remove artifacts, or compensate for temperature effects. If the analysis 103 of the image information indicates that the biometric object is not part of a live individual, then an activity requested (including activities that are merely desired, such as access to an application or authorization of a purchase) may be denied 150 . For example, if the individual requests access to a facility, computer database, or mobile device application and the information analysis indicates that the biometric object was not part of a live individual, then the individual may not be permitted to gain access to the facility, database or application, as the case may be.
  • the individual may be allowed 153 to gain access to the facility, database, or application, as the case may be.
  • identification, verification or authentication of the user may be required in addition to an indication of liveness before access may be granted or a requested activity is allowed.
  • FIGS. 2-3 illustrate various systems and methods that may be used to evaluate an image of a biometric object.
  • the systems and methods include determining whether an image of a biometric object is of a live individual.
  • FIG. 2 shows a schematic of an exemplary biometric authentication method, which may be used in conjunction with one or more additional systems for evaluating fingerprints.
  • the system can be embodied as separate layers for filtering the image of a detected biometric object.
  • the layers can be software-based filters that are used to improve efficiency, accuracy, security, and/or usability.
  • the biometric authentication method can include a skin-like object detection layer 210 , which can be used to filter one or more of the following features: air and object; skin and non-skin; finger and stylus.
  • the skin-like object detection layer 210 can determine whether a detected acoustic impendence is within a particular range.
  • ultrasound waves propagate through mediums of varying acoustic impedance. Therefore reflection and transmission coefficients of ultrasound can be used to obtain images.
  • Acoustic impedance can be the product of biometric-density and phase velocity, respectively. As such, the acoustic impedance values may be different for ridges and valleys in the biometric. In some implementations, the variation between acoustic impedance in the ridges and valleys may be determinative, and thus it may not be necessary to detect a quantitative acoustic impedance.
  • the biometric authentication method can include a finger detection layer 220 , which can be used to filter one or more of the following features: fingerprint and non-fingerprint pattern; ridge, ridge flow, and fingerprint edge.
  • the finger detection layer 220 may involve a determination of whether the object includes typical fingerprint patterns (e.g., ridges and valleys). Alternatively, or additionally, the finger detection layer 220 may involve determining whether an object has an acoustic impedance that is within an acoustic impedance range corresponding to that of skin. In some examples, determining whether the object is a finger may involve distinguishing a finger from another body part. Determining whether the object is a finger may involve determining that a non-finger object is a stylus.
  • the finger detection layer 220 may provide a finger/non-finger output signal indicating whether a finger is on the fingerprint sensor.
  • the finger detection layer 220 may provide an acoustic impedance output signal indicating whether the acoustic impedance of an object on the fingerprint sensor is within a predetermined range.
  • the finger detection layer 220 may provide a low-level fraud output signal indicating whether an object on the fingerprint sensor may be a fraud.
  • the finger detection layer 220 may provide a stylus/non-stylus output signal indicating whether a tip of a stylus is on the sensor.
  • the biometric authentication method can include an image qualification layer 230 to measure the quality of a biometric image, and in some instances, invoke a routine to improve the quality of the image.
  • the image qualification layer 230 can be used to perform one or more of the following features: filter low-quality fingerprint images, provide an image quality score, and enhance or re-acquire an image.
  • the image qualification layer 230 may provide an image quality output signal indicating whether an image quality is above a threshold.
  • the biometric authentication method can include a template qualification layer 240 , which may involve quality scoring of fingerprint features and/or templates, and may involve providing an indication of fingerprint matching confidence.
  • the template qualification layer 240 can be used to perform one or more of the following features: provide a feature quality score, a template quality score, and match confidence prediction.
  • the image qualification layer 230 may provide and/or store a feature quality output signal indicating a level of quality for a feature.
  • the image qualification layer 230 may provide and/or store a template quality output signal indicating a level of quality of a template.
  • the image qualification layer 230 may provide a recognition confidence level output signal indicating a level of confidence that a template could be matched, or whether the recognition confidence level is above a predetermined threshold.
  • One or more of these output signals may be stored, sent to or otherwise provided to various software applications running on a mobile device.
  • the biometric authentication method can include a matching/enrollment layer 250 , which can be used to perform pattern or minutiae matching.
  • a fingerprint template can be generated based, at least in part, on extracted features of a biometric object.
  • fingerprint template information can be generated that may include the type, size, location and/or orientation of fingerprint patterns and/or fingerprint minutiae.
  • the biometric authentication method can include a liveness detection layer 260 , which can be used to determine liveness.
  • the liveness detection layer may be used to provide a liveness detection output score or make a determination of liveness.
  • the liveness detection layer can be specifically be embodied as one or more of the following methods and systems described further herein.
  • FIG. 3 depicts a first implementation of a method of detecting liveness.
  • the method can include obtaining an image of a biometric object.
  • the image may be a visual representation of the biometric object and/or a data representation of the biometric object. Although not visually recognizable as an image of the biometric object, the data representation may nonetheless contain identifying data corresponding to the biometric object.
  • only a single ultrasonic image of a biometric object is obtained.
  • An image block can be formed from the ultrasonic image.
  • the image block may be selected to include relevant areas of the image, such as, areas containing a fingerprint or fingerprint features.
  • the image block may be subdivided into a plurality of sample blocks. The sample blocks may be of the same, different, or overlapping areas.
  • certain areas of the image block may be divided into larger or smaller sample blocks based on the content of the image block.
  • Feature vectors in a spatial domain and a frequency domain can be extracted from the image block.
  • the feature vectors are extracted from each of the plurality of sample blocks.
  • the feature vectors from each of the plurality of sample blocks can be compared to a classification model. The comparison can be used to make a classification or determination of liveness.
  • FIG. 4 illustrates generalized flows for fingerprint enrollment and/or authentication followed by a liveness determination stage according to aspects of the present disclosure.
  • block 400 describes the enrollment process and block 420 describes the verification/authentication process.
  • an acquired image may be processed to generate a template (e.g. template information, template data, biometric reference data, or reference) that may be stored in a local or external database 410 .
  • a reference may include one or more templates, models, or raw images.
  • the enrollment process may include image acquisition 402 , image processing 404 , feature extraction 406 , template generation 408 , and data storage in a database 410 .
  • the pattern types may be stored as part of a template associated with an object such as a finger.
  • pattern classifications may be determined during enrollment.
  • the pattern types and pattern classifications may be used in part for authentication or verification.
  • the pattern types and/or pattern classifications for a particular user may be updated.
  • the updated pattern types and/or pattern classifications may be stored in a modified template.
  • the verification/authentication process in block 420 may include image acquisition 422 , image processing 424 , feature extraction 426 , template generation 428 , fingerprint matching 430 using information stored in the database 410 , and match determination 432 to determine and provide a match output signal 434 .
  • each acquired image may be processed to generate a template; the generated templates may be used for matching.
  • the fingerprint verification/authentication block 420 may provide a match output signal 434 indicating whether a match has occurred.
  • One or more liveness determination stages 440 may perform various temporal analyses and/or expected biometric analyses to determine whether a finger is real or fake and whether the finger is dead or alive.
  • a liveness output signal 442 may be provided that indicates a liveness determination.
  • a liveness determination 440 to provide a liveness output signal 442 may be made during the fingerprint enrollment block 400 .
  • the liveness output signal 442 may be provided to determine whether to enroll a user.
  • the liveness output signal 442 may be provided to determine whether to verify, identify, or authenticate a user.
  • the liveness output signal 442 may be combined with a match output signal 434 to determine whether to authenticate or verify a user.
  • a positive match output signal 434 and a positive liveness output signal 442 may be combined to allow access or to allow a requested activity.
  • the liveness output signal 442 may be generated and provided to a software application or an application running on a mobile or non-mobile device.
  • pattern types may be determined during enrollment.
  • FIG. 5B is a fingerprint image showing its ridge-valley mask.
  • Feature vectors may be extracted from the image that can be used in determining liveness.
  • raw textural information such as histogram-based features (“histogram-features”)
  • histogram-features can be extracted for both ridge and valley from an image, such as an ultrasonic image.
  • information may be obtained about the ultrasonic impedance characteristics of ridges and valleys of a finger, for example, by scanning the fingerprint with an ultrasonic scanner.
  • Histogram features may be related to an entire image block, a subset of the image block, or along a vector superimposed on the image. The obtained information may be statistically analyzed to determine histogram-features corresponding to the obtained information.
  • ratio-features may be used to obtain ratio-features. These ratio-features are another example of a vector feature that may be used to determine whether the imaged object is live or not. In many instances, ratio-features may be a ratio of individual features between ridge and valley. Ratio-features may be related to the entire image block, a subset of the image block, between one or more ridge/valley combinations, or along a vector superimposed on the image.
  • ratio-features can be measurements of the image statistics of ridges and valleys of the imaged object.
  • ratio-features from real fingerprint images and ratio-features from fake fingerprint images each may show a distribution in feature domain, which can be used to determine whether the imaged object was real or fake.
  • the method can take into account the ridge and valley of an imaged object simultaneously, such as by extracting features that take into account a ridge and valley simultaneously.
  • cross-ridge-valley-features may include co-occurrence characteristics of ridge and valley, as well as the transition pattern of ridge-to-valley and valley-to-ridge in terms of gray scale values representing acoustic impedance of the imaged object.
  • Cross-ridge-valley-features may be displayed, analyzed, or compared in spatial and/or frequency domains, such as one or more of the following: ridge and valley separation in gray scale values, ridge/valley width, ridge-to-valley transition rate, valley-to-ridge transition rate, and the frequency spectrum of the fast Fourier transform of the gray scale profile along a certain direction.
  • the direction can be defined using a vector, such as the direction that is perpendicular to the ridge flow of the biometric object.
  • Such cross-ridge-valley-features may be useful in identifying co-occurrence characteristics of the ridge and valley as well as the changing pattern of ridge-to-valley and valley-to-ridge. This may be carried out in terms of gray scale levels in the image.
  • the grey level profile in the direction perpendicular to the ridge flow may be analyzed, and various features extracted to characterize the co-occurrence of the ridge and valley, in both spatial and frequency domains. For instance, one or more of the following features may be extracted: ridge and valley separation, width ratio, ridge and valley changing rate, power spectrum of a fast Fourier transform. Additional features of the co-occurrence of the ridge and valley, in both spatial and frequency domains, may additionally be used.
  • Methods that identify textural components can be more difficult to spoof than methods that merely identify geometric components.
  • a fake biometric will have a noticeably different gray level distribution across ridges and valleys as compared to a real biometric. The gray level distribution across ridges and valleys can therefore be used to more reliably distinguish between a spoof image and real image of a biometric object.
  • FIGS. 6A and 6B provide examples of a cross-ridge-valley measurement.
  • a cross-ridge-valley measurement the two-dimensional spatial pattern consistency of the ridge and valley of an imaged object can be extracted.
  • FIG. 6B is a plot showing the brightness profile along the line segment 500 shown in FIG. 6A , which crosses several fingerprint ridge/valley pairs.
  • the x-axis units represent the pixel location along the selected direction, and the y-axis is the gray scale values of each pixel along the line segment 500 .
  • the brightness profile in FIG. 6B is therefore a characterization of the ridge/valley pairs along the line segment 500 of the image of FIG. 6A .
  • a measure of the image brightness of ridges and a measure of the image brightness of the valleys can be used to determine liveness.
  • a fingerprint of a live being may produce image patches, e.g. a portion of the image, that produce a more limited value-range for CRV BT than image patches provided by a fake fingerprint.
  • a fingerprint image patch of a live being can produce larger values for CRV freq than image patches provided by a fake fingerprint.
  • a fake fingerprint can generate images with higher frequency noise, and a less clear ridge/valley pattern.
  • the energy of a fake image in the frequency domain can distribute to more frequencies than a genuine image—thereby making the quality value for the genuine image higher than a fake image.
  • one or more image quality measurements may be included in a feature vector.
  • the image quality measurement can be one or more of the following: the number of connected components in the fingerprint image, the smoothness of the components, the entropy of the orientation distribution, and the strength (i.e. the clarity/sharpness) of the ridge and valley.
  • OS ⁇ (i,j) ⁇ (k,l) 1 ⁇ abs(Ori( i,j ) ⁇ Ori( i ⁇ k,j ⁇ l ))/(Ori( i,j )+Ori( i ⁇ kj ⁇ l ) Ridge orientations smoothness
  • a fake fingerprint usually presents poorer image qualities. This can be because an image of a fake fingerprint can exhibit more variation in orientaions in the local window, thereby resulting in a smaller OS value. In contrast, images of genuine fingeringrint can produce more uniform orientation in the local window, thereby resulting in a larger OS value than a fake fingerprint.
  • OCL ( H ( i _max)+ ⁇ ( H ( i _max ⁇ 1)+ H ( i _max+1)))/ ⁇ H ( i ) Ridge orientation certainty level
  • Good quality fingerprint patches can have a larger value for this feature because of their clearer ridge flow information.
  • genuine fingerprint images of good quality can have a high response to only one or two Gabor filters because it may have one or two major orientations.
  • Fake fingerprint images of poor quality can have a higher response to all Gabor filters. Therefore, the standard deviation of all of the filter response for a genuine image can be higher than fake images.
  • Yet another methodology may be based on capturing special image patterns generated during the imaging process by different finger materials (e.g. skin and non-skin; living and non-living).
  • this group of features may characterize the image information contributed by an actual fingerprint (e.g. the fingerprint texture shown in the image) and the detailed information, which could be from the difference in the ultrasound impedance of sensing material (skin vs. non-skin), or the imaging process noise.
  • skin and non-skin materials respond to the ultrasound signal is different, and the special patterns presented in the images are different in terms of these feature measurements.
  • wavelet-based features are used to determine a measure of the wavelet packet decomposition, which would provide an approximation and a detail component of the original image.
  • the features from the approximation component could include:
  • Eg ⁇ ( I ( i,j ) ⁇ )( I ( i,j ) ⁇ )
  • Ep ⁇ ( I ( i,j ) ⁇ )log
  • s ⁇ ( I ( i,j ) ⁇ )*( I ( i,j ) ⁇ )/( I ( i,j ) ⁇ )/( ⁇ * ⁇ * ⁇ )
  • f ⁇ ( I ( i,j ) ⁇ )*( I ( i,j ) ⁇ )/( I ( i,j ) ⁇ )*( I ( i,j ) ⁇ )/( ⁇ * ⁇ * ⁇ * ⁇ )
  • a subject-specific scheme may be utilized.
  • a classifier is trained for each individual user such that one user is authenticated as to liveness using a particular set of features and parameters that are different from those used to authenticate a different user with regard to liveness.
  • the classifier may be programmed to identify the set of features and parameters that are best for each user.
  • FIGS. 7-9 illustrate various methods that may be used to determine liveness of a biometric object.
  • FIG. 7 provides a method of training 600 that provides baseline liveness detection model.
  • a baseline liveness detection model can be trained using fingerprint data collected before the system is deployed for use.
  • a first step 601 may include collecting large amounts of biometric data (e.g. fingerprint images) from both real and fake biometrics.
  • a second step 602 may include extracting features from of a biometric image (e.g. one or more fingerprint images). The second step 602 can include having one or more block samples go through a feature extraction module.
  • a third step 603 can include extracting one or more image features of the biometric image.
  • the features can include one or more feature vectors, as further described herein.
  • a fourth step 604 can include taking feature vectors from real and fake images as positive and negative samples.
  • the feature vectors may be sent to a training engine, which could be a machine learning algorithm.
  • the training engine can be used to create a classification model 605 .
  • the classification model may serve as a baseline model.
  • the baseline model may be used with a system or method for determining liveness.
  • FIG. 8 provides a method 610 of enrollment and refinement of a liveness detection model.
  • the baseline model can be updated so that the model is tailored to the new user. In this manner, a higher level of accuracy of determining liveness may be available to the new user.
  • one or more images of a biometric object of a new user can be obtained 611 .
  • One or more feature vectors may be extracted 612 from the one or more images.
  • An incremental training algorithm may be used 613 with the extracted feature vectors, along with the baseline model in order to create a new classification model 614 .
  • the new classification model 614 can therefore be tailored towards the liveness detection for a particular user.
  • FIG. 9 provides an implementation of a method 620 of determining liveness.
  • the method 620 may be used, for example when a user is to be authenticated.
  • the method 620 can include capturing 621 one or more images of a biometric object of the user, for example the user's fingerprint. In one particular example, a single ultrasonic image of the biometric object can be obtained.
  • One or more feature vectors can be extracted 622 from the image.
  • a liveness detection engine (“LivDet Engine”) may process 623 the extracted feature vectors, along with a user-specific model in order to provide a liveness measurement. Based upon the liveness rating, a decision regarding liveness can be made.
  • each feature vector can represent one sample of data. Consequently, each feature vector may be associated with a liveness measurement or decision. Decisions from the feature vectors can be fused 624 using one or more algorithms, including those algorithms described herein. A final decision, for example an image-level decision, can be reach and outputted.
  • an oversampling strategy may be adopted so as to extract a large amount of data. Oversampling may be used to make a robust decision, despite issues related to a fingerprint scan, including possible sensor noise and poor fingerprint qualities.
  • an “image block” to be a scan of a fingerprint like that shown in FIG. 10 .
  • the image block may be subdivided into sample blocks. There may be overlapping between sample blocks. In FIG. 10 , multiple colors are used to better distinguish between sample blocks, and it should be noted that overlapping sample blocks that are drawn later are shown covering the earlier rendered sample block's color.
  • Each sample block in FIG. 10 represents 8 ⁇ 8 pixels.
  • sample blocks are defined only in areas containing fingerprint data.
  • the number of pixels in the image block may depend on the sensor size and resolution. For a typical image block, there may be 50 to 200 sample blocks that can be used, and the actual number of usable sample blocks often depends on the size and quality of the fingerprint image.
  • each sample block For each sample block, one or more of the feature vectors described above may be extracted. So, for a particular image block, many types of features may be obtained for each sample block, and since there are many sample blocks in an image block, each type of feature may have a large number of data, thereby facilitating the production of a statistically significant sample.
  • Another advantage of using multi-block overlapping sampling approach can be that the impact of a poorly imaged fingerprint area, which could be caused by worn fingerprints or defective sensor patches, may be minimized.
  • each sample block contributes one feature vector, and based on that feature vector a determination as to liveness is made with respect to that sample block.
  • a determination as to liveness is made with respect to that sample block.
  • each sample block can “vote” equally as to whether a real fingerprint or a fake fingerprint is indicated by its feature vector. If the number of sample blocks indicating a real fingerprint outnumbers the number of sample blocks indicating a fake fingerprint, then a real fingerprint will be indicated. However, if the number of sample blocks indicating a real fingerprint does not outnumber the number of sample blocks indicating a fake fingerprint, then a fake fingerprint will be indicated.
  • each sample block “votes”, but some “votes” can be weighted more than other “votes”.
  • the weight for a particular sample block's vote could be derived from an image quality measurement of this block. Or, the weight could come from the probability of a sample being real or spoof fingerprint estimated by a learned model.
  • Such a decision-making model can allow for easy handling of situations in which multiple fingerprint images are available.
  • sample blocks from all the images could be placed (in effect) in a sample pool, and a decision regarding liveness made based on that sample pool.
  • individual image-level decisions could be fused in order to make a final decision regarding liveness.
  • a method of determining liveness 800 may include one or more machine learning algorithms to interpret data derived from the methods of determining liveness described above, including algorithms such as support vector machine (SVM), Random forest, and extreme learning machine (ELM).
  • SVM support vector machine
  • ELM extreme learning machine
  • an image of a biometric may be divided into sample blocks 810 .
  • the sample blocks may be distinguished 820 between real and fake.
  • the machine learning algorithm can be trained 830 based upon the real and fake sample blocks.
  • the output of the training may be used to create 840 a a classification model, or to improve 840 b an existing classification model, against which future sample blocks would be compared.
  • the classification model may be used in a comparison to determine whether an image is from a real or a fake biometric object.
  • a method of determining liveness 900 may include an adaptive liveness detection model.
  • the method 900 can include obtaining 910 a new image from an enrolled user after the user enrolls onto a device.
  • the new image can be used to update or tune 930 a previously trained classification model.
  • the update or tune may be used to better tune the trained classification model to a specific individual, or more generally to improve a classification model or group of classification models to distinguish a real fingerprint from a fake fingerprint.
  • An adaptive liveness detection model can advantageously updating the model as a biometric object (e.g. fingerprint) changes with time (e.g. due to frequent use, which leads to wearing out, or natural aging of the person). By updating the classification model according to a change in the biometric object, the accuracy of the system can be maintained—or possibly improved over time.
  • the method 900 may optionally determine 920 whether the new image meets one or more constraints before updating or tuning a previously trained classification model.
  • the method 900 may include constraints such as requiring that a biometric pass a matching process for (e.g. one or more biometric scans and authentication processes), and/or the fingerprint may need to pass a minimum threshold liveness determination in order to allow the classification model to update. These constraints can be used to avoid spoofing an adaptive liveness detection model to update its classification model according to a fake biometric image.
  • a method of detecting liveness 1000 may include the step of obtaining 1010 a single ultrasonic image of a biometric object.
  • the single ultrasonic image may be subdivided into a plurality of overlapping sample blocks.
  • the feature vectors may be extracted 1030 in a spatial domain and a frequency domain from each of the plurality of sample blocks.
  • the feature vectors may be compared 1040 from each of the plurality of sample blocks to a classification model.
  • the classification model may be altered 1060 based on the comparison of the feature vectors to the classification model.
  • the classification model may be altered 1050 based on the single ultrasonic image of the biometric object.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Collating Specific Patterns (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)
  • Quality & Reliability (AREA)
  • Image Input (AREA)
US14/845,174 2014-09-05 2015-09-03 Image-based liveness detection for ultrasonic fingerprints Active US9633269B2 (en)

Priority Applications (13)

Application Number Priority Date Filing Date Title
US14/845,174 US9633269B2 (en) 2014-09-05 2015-09-03 Image-based liveness detection for ultrasonic fingerprints
CA2957218A CA2957218C (en) 2014-09-05 2015-09-04 Multi-stage liveness determination
CN201580045104.7A CN106663198A (zh) 2014-09-05 2015-09-04 用于超声波指纹的基于图像的活性检测
KR1020177005953A KR102467916B1 (ko) 2014-09-05 2015-09-04 라이브니스 결정을 위한 방법, 시스템, 및 비-일시적 컴퓨터 판독가능 저장 매체
PCT/US2015/048581 WO2016037077A1 (en) 2014-09-05 2015-09-04 Image-based liveness detection for ultrasonic fingerprints
EP15772060.8A EP3189470A1 (en) 2014-09-05 2015-09-04 Image-based liveness detection for ultrasonic fingerprints
JP2017511746A JP6656228B2 (ja) 2014-09-05 2015-09-04 多段式生存性判定
BR112017004379-3A BR112017004379B1 (pt) 2014-09-05 2015-09-04 Determinação de vivacidade em multi-estágio
PCT/US2015/048583 WO2016037078A1 (en) 2014-09-05 2015-09-04 Multi-stage liveness determination
CN201580046844.2A CN106796650B (zh) 2014-09-05 2015-09-04 多阶段活性确定
EP15772061.6A EP3189471B1 (en) 2014-09-05 2015-09-04 Multi-stage liveness determination
KR1020227039636A KR102636547B1 (ko) 2014-09-05 2015-09-04 라이브니스 결정을 위한 방법, 시스템, 및 비-일시적 컴퓨터 판독가능 저장 매체
AU2015311785A AU2015311785B9 (en) 2014-09-05 2015-09-04 Multi-stage liveness determination

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US201462046744P 2014-09-05 2014-09-05
US201562151983P 2015-04-23 2015-04-23
US14/845,174 US9633269B2 (en) 2014-09-05 2015-09-03 Image-based liveness detection for ultrasonic fingerprints

Publications (2)

Publication Number Publication Date
US20160070968A1 US20160070968A1 (en) 2016-03-10
US9633269B2 true US9633269B2 (en) 2017-04-25

Family

ID=55437785

Family Applications (3)

Application Number Title Priority Date Filing Date
US14/845,174 Active US9633269B2 (en) 2014-09-05 2015-09-03 Image-based liveness detection for ultrasonic fingerprints
US14/845,149 Active US9639765B2 (en) 2014-09-05 2015-09-03 Multi-stage liveness determination
US15/472,173 Active US9953233B2 (en) 2014-09-05 2017-03-28 Multi-stage liveness determination

Family Applications After (2)

Application Number Title Priority Date Filing Date
US14/845,149 Active US9639765B2 (en) 2014-09-05 2015-09-03 Multi-stage liveness determination
US15/472,173 Active US9953233B2 (en) 2014-09-05 2017-03-28 Multi-stage liveness determination

Country Status (9)

Country Link
US (3) US9633269B2 (enExample)
EP (2) EP3189471B1 (enExample)
JP (1) JP6656228B2 (enExample)
KR (2) KR102467916B1 (enExample)
CN (2) CN106796650B (enExample)
AU (1) AU2015311785B9 (enExample)
BR (1) BR112017004379B1 (enExample)
CA (1) CA2957218C (enExample)
WO (2) WO2016037078A1 (enExample)

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9953233B2 (en) 2014-09-05 2018-04-24 Qualcomm Incorporated Multi-stage liveness determination
US10135822B2 (en) 2017-03-21 2018-11-20 YouaretheID, LLC Biometric authentication of individuals utilizing characteristics of bone and blood vessel structures
US10325137B2 (en) * 2006-04-26 2019-06-18 Aware, Inc. Fingerprint preview quality and segmentation
US10685204B2 (en) 2018-07-27 2020-06-16 Qualcomm Incorporated Biometric age estimation via ultrasonic imaging
US10880303B2 (en) 2017-03-21 2020-12-29 Global E-Dentity, Inc. Real-time COVID-19 outbreak identification with non-invasive, internal imaging for dual biometric authentication and biometric health monitoring
US11335116B2 (en) 2017-07-20 2022-05-17 Seung Jin Lee Apparatus and method for identifying fake fingerprint by using impedance
US11374929B2 (en) 2017-03-21 2022-06-28 Global E-Dentity, Inc. Biometric authentication for an augmented reality or a virtual reality device
US11468712B2 (en) 2020-01-09 2022-10-11 AuthenX Inc. Liveness detection apparatus, system and method
US11582032B2 (en) 2019-09-25 2023-02-14 Amod Ashok Dange System and method for sharing user preferences without having the user reveal their identity
US20230101088A1 (en) * 2021-09-28 2023-03-30 Samsung Display Co., Ltd. Method and apparatus for restoring image of fingerprint sensor
US20230360653A1 (en) * 2022-05-04 2023-11-09 Board Of Trustees Of Michigan State University Speaker Verification System Using Ultrasound Energy In Human Speech
US12028347B2 (en) 2019-09-25 2024-07-02 Amod Ashok Dange System and method for enabling a user to obtain authenticated access to an application using a biometric combination lock
US12026247B2 (en) 2019-09-25 2024-07-02 Amod Ashok Dange System and method for enabling a user to create an account on an application or login into the application without having the user reveal their identity
US12051267B2 (en) 2020-04-29 2024-07-30 Qualcomm Incorporated Fingerprint sensor with detection of latent fingerprints
US12072963B2 (en) 2019-09-25 2024-08-27 Amod Ashok Dange System and method for affixing a signature using biometric authentication
US12079367B2 (en) 2019-09-25 2024-09-03 Amod Ashok Dange System and method for enabling social network users to grant their connections granular visibility into their personal data without granting the same visibility to the network operator

Families Citing this family (97)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170372124A1 (en) * 2014-12-24 2017-12-28 Sciometrics Llc Unobtrusive identity matcher: a tool for real-time verification of identity
US10387704B2 (en) * 2015-06-29 2019-08-20 Qualcomm Incorporated Method and apparatus for enabling the touchscreen display of a mobile device
US10254881B2 (en) 2015-06-29 2019-04-09 Qualcomm Incorporated Ultrasonic touch sensor-based virtual button
US9726755B2 (en) * 2015-09-23 2017-08-08 Qualcomm Incorporated Spoof detection by ultrasonic subdermal probe
US10140534B2 (en) 2015-09-26 2018-11-27 Qualcomm Incorporated Ultrasonic imaging devices and methods
US10262188B2 (en) * 2016-02-15 2019-04-16 Qualcomm Incorporated Liveness and spoof detection for ultrasonic fingerprint sensors
CN107091704B (zh) * 2016-02-17 2020-10-09 北京小米移动软件有限公司 压力检测方法和装置
FR3049090B1 (fr) * 2016-03-21 2021-06-25 Sebastien Jean Serge Dupont Dispositif d'authentification biometrique adaptatif par echographie, photographies en lumiere visible de contraste et infrarouge, sans divulgation, a travers un reseau informatique decentralise
US9961178B2 (en) * 2016-03-24 2018-05-01 Motorola Mobility Llc Embedded active matrix organic light emitting diode (AMOLED) fingerprint sensor
CN109074482A (zh) * 2016-05-03 2018-12-21 精确生物识别公司 用于生物活性检测的用户特定分类器
US10315222B2 (en) 2016-05-04 2019-06-11 Invensense, Inc. Two-dimensional array of CMOS control elements
US10445547B2 (en) 2016-05-04 2019-10-15 Invensense, Inc. Device mountable packaging of ultrasonic transducers
US10235551B2 (en) * 2016-05-06 2019-03-19 Qualcomm Incorporated Biometric system with photoacoustic imaging
US10562070B2 (en) 2016-05-10 2020-02-18 Invensense, Inc. Receive operation of an ultrasonic sensor
US10706835B2 (en) 2016-05-10 2020-07-07 Invensense, Inc. Transmit beamforming of a two-dimensional array of ultrasonic transducers
US10441975B2 (en) 2016-05-10 2019-10-15 Invensense, Inc. Supplemental sensor modes and systems for ultrasonic transducers
US10539539B2 (en) 2016-05-10 2020-01-21 Invensense, Inc. Operation of an ultrasonic sensor
US11673165B2 (en) 2016-05-10 2023-06-13 Invensense, Inc. Ultrasonic transducer operable in a surface acoustic wave (SAW) mode
US10452887B2 (en) 2016-05-10 2019-10-22 Invensense, Inc. Operating a fingerprint sensor comprised of ultrasonic transducers
TWI591545B (zh) * 2016-06-30 2017-07-11 指紋影像偵測裝置及其方法
US10460144B2 (en) 2016-07-20 2019-10-29 Cypress Semiconductor Corporation Non-finger object rejection for fingerprint sensors
US10599911B2 (en) 2016-07-20 2020-03-24 Cypress Semiconductor Corporation Anti-spoofing protection for fingerprint controllers
JP6792986B2 (ja) * 2016-09-27 2020-12-02 株式会社日立製作所 生体認証装置
WO2018082011A1 (zh) * 2016-11-04 2018-05-11 深圳市汇顶科技股份有限公司 活体指纹识别方法及装置
US10176362B1 (en) * 2016-11-10 2019-01-08 Synaptics Incorporated Systems and methods for a gradient-based metric for spoof detection
US10430638B2 (en) * 2016-11-10 2019-10-01 Synaptics Incorporated Systems and methods for spoof detection relative to a template instead of on an absolute scale
US10262187B1 (en) * 2016-11-10 2019-04-16 Synaptics Incorporated Systems and methods for spoof detection based on local binary patterns
US9946914B1 (en) * 2016-11-18 2018-04-17 Qualcomm Incorporated Liveness detection via ultrasonic ridge-valley tomography
US10552662B2 (en) * 2016-12-30 2020-02-04 Beyond Time Investments Limited Optical identification method
CN108304759A (zh) * 2017-01-11 2018-07-20 神盾股份有限公司 识别手指的方法和电子装置
KR102459852B1 (ko) * 2017-02-08 2022-10-27 삼성전자주식회사 지문 인식을 위한 후보 지문 이미지를 선정하는 장치 및 방법
US10515255B2 (en) * 2017-03-24 2019-12-24 Qualcomm Incorporated Fingerprint sensor with bioimpedance indicator
US10552658B2 (en) 2017-03-24 2020-02-04 Qualcomm Incorporated Biometric sensor with finger-force navigation
US10438040B2 (en) 2017-03-24 2019-10-08 Qualcomm Incorporated Multi-functional ultrasonic fingerprint sensor
US10515199B2 (en) 2017-04-19 2019-12-24 Qualcomm Incorporated Systems and methods for facial authentication
EP3631665A4 (en) * 2017-05-23 2021-03-03 Fingerprint Cards AB METHOD AND ELECTRONIC DEVICE FOR AUTHENTICATING A USER
US10474862B2 (en) 2017-06-01 2019-11-12 Invensense, Inc. Image generation in an electronic device using ultrasonic transducers
SE1750720A1 (en) 2017-06-07 2018-12-08 Fingerprint Cards Ab Fingerprint authentication method and system for rejecting spoof attempts
CN107609463B (zh) * 2017-07-20 2021-11-23 百度在线网络技术（北京）有限公司 活体检测方法、装置、设备及存储介质
KR102724372B1 (ko) * 2017-08-09 2024-10-30 더 보드 오브 트러스티즈 오프 더 리랜드 스탠포드 쥬니어 유니버시티 인터랙티브 생체 터치 스캐너
CN111201537B (zh) 2017-10-18 2023-11-17 指纹卡安娜卡敦知识产权有限公司 在指纹分析中通过机器学习来区分活体手指与欺骗手指
KR102361752B1 (ko) * 2017-11-24 2022-02-11 한국전자기술연구원 초음파 도플러 효과를 이용하여 생체를 판별하는 지정맥 인식 방법 및 장치
US10936841B2 (en) 2017-12-01 2021-03-02 Invensense, Inc. Darkfield tracking
US10997388B2 (en) 2017-12-01 2021-05-04 Invensense, Inc. Darkfield contamination detection
US10783346B2 (en) * 2017-12-11 2020-09-22 Invensense, Inc. Enhancing quality of a fingerprint image
US11151355B2 (en) 2018-01-24 2021-10-19 Invensense, Inc. Generation of an estimated fingerprint
US11010589B1 (en) 2018-03-16 2021-05-18 Synaptics Incorporated Defensive measures for residue re-imaging
US11120247B2 (en) 2018-03-16 2021-09-14 Synaptics Incorporated Defensive measures for residue re-imaging
US10755067B2 (en) 2018-03-22 2020-08-25 Invensense, Inc. Operating a fingerprint sensor comprised of ultrasonic transducers
WO2019216499A1 (ko) * 2018-05-08 2019-11-14 엘지전자 주식회사 전자 장치 및 그 제어 방법
WO2019216498A1 (en) * 2018-05-08 2019-11-14 Lg Electronics Inc. Electronic device and method for controlling the same
US10803159B2 (en) 2018-05-14 2020-10-13 Lg Electronics Inc. Electronic device and method for controlling the same
WO2019238251A1 (en) * 2018-06-13 2019-12-19 Veridas Digital Authentication Solutions, S.L. Authenticating an identity of a person
CN109492551B (zh) * 2018-10-25 2023-03-24 腾讯科技（深圳）有限公司 活体检测方法、装置及应用活体检测方法的相关系统
US11023756B2 (en) * 2018-10-26 2021-06-01 Advanced New Technologies Co., Ltd. Spoof detection using iris images
US10706306B2 (en) * 2018-12-03 2020-07-07 Qualcomm Incorporated Spoof and liveness detection via pyroelectric indications
US10936843B2 (en) 2018-12-28 2021-03-02 Invensense, Inc. Segmented image acquisition
EP3915045B1 (en) * 2019-02-21 2025-10-15 Next Biometrics Group ASA Method of detecting replay attacks in a fingerprint sensor system
US11216638B2 (en) * 2019-05-09 2022-01-04 Qeexo, Co. Method and system to prevent identity theft for fingerprint recognition enabled touch screen devices
CN110287866B (zh) * 2019-06-24 2021-10-26 Oppo广东移动通信有限公司 超声波指纹识别方法、装置、存储介质及电子设备
WO2020263875A1 (en) * 2019-06-24 2020-12-30 Invensense, Inc. Fake finger detection using ridge features
US11216681B2 (en) 2019-06-25 2022-01-04 Invensense, Inc. Fake finger detection based on transient features
US11216632B2 (en) 2019-07-17 2022-01-04 Invensense, Inc. Ultrasonic fingerprint sensor with a contact layer of non-uniform thickness
US11176345B2 (en) 2019-07-17 2021-11-16 Invensense, Inc. Ultrasonic fingerprint sensor with a contact layer of non-uniform thickness
US20210056276A1 (en) * 2019-08-19 2021-02-25 Qualcomm Incorporated Apparatus and method for improved biometric sensing
US11232549B2 (en) 2019-08-23 2022-01-25 Invensense, Inc. Adapting a quality threshold for a fingerprint image
CN110579271B (zh) * 2019-09-24 2022-04-19 成都大超科技有限公司 超声波指纹识别模组及电子设备
WO2021061971A1 (en) * 2019-09-27 2021-04-01 Butterfly Network, Inc. Methods and apparatus for configuring an ultrasound system with imaging parameter values
US11392789B2 (en) 2019-10-21 2022-07-19 Invensense, Inc. Fingerprint authentication using a synthetic enrollment image
TWI712391B (zh) * 2019-11-19 2020-12-11 台灣立訊精密有限公司 動物體能追蹤裝置及其系統
CN112989888B (zh) * 2019-12-17 2024-06-07 华为技术有限公司 一种指纹防伪方法及电子设备
US11625947B1 (en) * 2020-03-03 2023-04-11 Amazon Technologies, Inc. System for detecting and mitigating fraudulent biometric input
EP4100176B1 (en) 2020-03-09 2024-10-09 InvenSense, Inc. Ultrasonic fingerprint sensor with a contact layer of non-uniform thickness
US11243300B2 (en) 2020-03-10 2022-02-08 Invensense, Inc. Operating a fingerprint sensor comprised of ultrasonic transducers and a presence sensor
US11328165B2 (en) 2020-04-24 2022-05-10 Invensense, Inc. Pressure-based activation of fingerprint spoof detection
CN111368811B (zh) * 2020-05-26 2020-09-18 腾讯科技（深圳）有限公司 活体检测方法、装置、设备及存储介质
TWI762053B (zh) * 2020-06-05 2022-04-21 神盾股份有限公司 屏下式指紋感測裝置
US11163970B1 (en) 2020-06-16 2021-11-02 Google Llc Optical fingerprint system with varying integration times across pixels
US11241173B2 (en) * 2020-07-09 2022-02-08 Mediatek Inc. Physiological monitoring systems and methods of estimating vital-sign data
US11995909B2 (en) 2020-07-17 2024-05-28 Tdk Corporation Multipath reflection correction
US12174295B2 (en) 2020-08-07 2024-12-24 Tdk Corporation Acoustic multipath correction
KR102850054B1 (ko) * 2020-12-10 2025-08-25 삼성전자주식회사 생체 정보의 도용 여부를 검출하는 방법 및 장치
TW202226015A (zh) 2020-12-28 2022-07-01 神盾股份有限公司 用於指紋認證的方法、設備和非瞬態計算機可讀存儲介質
US12321445B2 (en) * 2021-03-31 2025-06-03 Samsung Electronics Co., Ltd. Method and apparatus with biometric information spoof detection
US11385770B1 (en) 2021-04-21 2022-07-12 Qualcomm Incorporated User interfaces for single-handed mobile device control
FR3123739B1 (fr) * 2021-06-02 2024-02-02 Idemia Identity & Security France Procede et dispositif d’identification et/ou d’authentification biometrique
US12416807B2 (en) 2021-08-20 2025-09-16 Tdk Corporation Retinal projection display system
US12260050B2 (en) 2021-08-25 2025-03-25 Tdk Corporation Differential receive at an ultrasonic transducer
US11721132B1 (en) 2022-01-28 2023-08-08 Armatura Llc System and method for generating region of interests for palm liveness detection
US11941911B2 (en) 2022-01-28 2024-03-26 Armatura Llc System and method for detecting liveness of biometric information
US11688204B1 (en) * 2022-01-28 2023-06-27 Armatura Llc System and method for robust palm liveness detection using variations of images
US12367821B2 (en) 2022-10-28 2025-07-22 Google Llc Expediting fingerprint authentication via variable refresh rate control and display luminance compensation
US20240164648A1 (en) * 2022-11-21 2024-05-23 Qualcomm Incorporated Safety methods for devices configured to emit high-intensity light
US20240232306A1 (en) * 2023-01-11 2024-07-11 Qualcomm Incorporated Liveness detection for an electronic device
WO2024177787A1 (en) 2023-02-23 2024-08-29 Google Llc Enlarging active areas of displays in electronic devices
US20240346733A1 (en) * 2023-04-11 2024-10-17 Disney Enterprises, Inc. Graph simulation for facial micro features with dynamic animation
US12008836B2 (en) 2023-05-04 2024-06-11 Google Llc Spatially and temporally dynamic illumination for fingerprint authentication

Citations (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5456256A (en) 1993-11-04 1995-10-10 Ultra-Scan Corporation High resolution ultrasonic imaging apparatus and method
WO2001024700A1 (en) 1999-10-07 2001-04-12 Veridicom, Inc. Spoof detection for biometric sensing systems
US20030202687A1 (en) 2002-04-29 2003-10-30 Laurence Hamid Method for preventing false acceptance of latent fingerprint images
US20050084154A1 (en) * 2003-10-20 2005-04-21 Mingjing Li Integrated solution to digital image similarity searching
US20050259850A1 (en) 2003-08-26 2005-11-24 Toshishige Shimamura Surface shape recognition sensor
WO2006042144A2 (en) 2004-10-07 2006-04-20 Ultra-Scan Corporation Ultrasonic fingerprint scanning utilizing a plane wave
US20070030475A1 (en) 2003-04-04 2007-02-08 Lumidigm, Inc. White-light spectral biometric sensors
US20070092115A1 (en) 2005-10-26 2007-04-26 Usher David B Method and system for detecting biometric liveness
EP1835438A1 (en) 2006-03-13 2007-09-19 Fujitsu Ltd. Fingerprint authenticating system
US20070230754A1 (en) 2006-03-30 2007-10-04 Jain Anil K Level 3 features for fingerprint matching
US20080253625A1 (en) 2006-10-10 2008-10-16 Stephanie Schuckers Multi-Resolutional Texture Analysis Fingerprint Liveness Systems and Methods
US7505613B2 (en) 2005-07-12 2009-03-17 Atrua Technologies, Inc. System for and method of securing fingerprint biometric systems against fake-finger spoofing
US20090316963A1 (en) 2008-06-19 2009-12-24 Authentec, Inc. Software based method for finger spoof detection and related devices
US20100008552A1 (en) 2008-07-14 2010-01-14 Union Community Co., Ltd. Apparatus for distinguishing living body fingerprint and method thereof
US20100113952A1 (en) 2008-11-03 2010-05-06 Raguin Daniel H Apparatus and method for the identification of fake fingerprints
US7738681B1 (en) 2005-11-01 2010-06-15 Hewlett-Packard Development Company, L.P. Fingerprint and physical attribute detection
US7804984B2 (en) 2006-07-31 2010-09-28 Lumidigm, Inc. Spatial-spectral fingerprint spoof detection
US20110158535A1 (en) * 2009-12-24 2011-06-30 Canon Kabushiki Kaisha Image processing apparatus and image processing method
US20120263355A1 (en) * 2009-12-22 2012-10-18 Nec Corporation Fake finger determination device
US20130058544A1 (en) 2009-09-22 2013-03-07 Thales Method for Detecting a Fake Finger for Fingerprint Acquisition Software
US20130202182A1 (en) 2011-08-09 2013-08-08 Lumidigm, Inc. Methods and systems for estimating genetic characteristics from biometric measurements
WO2014012467A1 (zh) 2012-07-20 2014-01-23 中国人民解放军军事医学科学院毒物药物研究所 间二芳基苯胺类或吡啶胺类化合物、其制备方法及医药用途
US20140226879A1 (en) 2013-02-08 2014-08-14 Apple Inc. Fingerprint Imaging and Quality Characterization
US20140241596A1 (en) 2013-02-28 2014-08-28 The Regents Of The University Of California Apparatus and Method for Capturing a Vital Vascular Fingerprint
US20160070967A1 (en) 2014-09-05 2016-03-10 Qualcomm Incorporated Multi-Stage Liveness Determination

Family Cites Families (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ATE252344T1 (de) * 1996-04-10 2003-11-15 Infineon Technologies Ag Verfahren zur personenidentifikation
JP2002175529A (ja) * 2000-12-06 2002-06-21 Matsushita Electric Ind Co Ltd 個人識別装置
JP2003051012A (ja) * 2001-08-03 2003-02-21 Nec Corp ユーザ認証方法及びユーザ認証装置
KR101035667B1 (ko) * 2002-05-09 2011-05-19 소니 주식회사 생체 인식 패턴 검출 장치, 개인 인증 장치 및 방법
FR2849246B1 (fr) * 2002-12-20 2006-03-10 Sagem Procede de determination du caractere vivant d'un element porteur d'une empreinte digitale
EP1611541B1 (en) * 2003-04-04 2010-12-15 Lumidigm, Inc. Multispectral biometric sensor
US20070098235A1 (en) * 2005-09-29 2007-05-03 Uriel Halavee Age Verification
CN101396277A (zh) * 2007-09-26 2009-04-01 中国科学院声学研究所 一种超声波人脸识别方法和识别装置
CN100514353C (zh) * 2007-11-26 2009-07-15 清华大学 一种基于人脸生理性运动的活体检测方法及系统
WO2012013202A1 (en) * 2010-07-27 2012-02-02 Max-Planck-Gesellschaft zur Förderung der Wissenschaften e. V. Dual-mode microwave resonator device and method of electron spin resonance measurement
CN102034117B (zh) * 2010-12-17 2012-08-22 东软集团股份有限公司 一种图像分类方法和装置
JP5699845B2 (ja) * 2011-07-29 2015-04-15 富士通株式会社 生体情報処理装置、生体情報処理方法及び生体情報処理用コンピュータプログラム
CN102368291B (zh) * 2011-09-07 2013-02-13 常州蓝城信息科技有限公司 基于指纹图像隐形意识的个人身份验证系统
CN102622588B (zh) * 2012-03-08 2013-10-09 无锡中科奥森科技有限公司 双验证人脸防伪方法及装置
US9396382B2 (en) * 2012-08-17 2016-07-19 Flashscan3D, Llc System and method for a biometric image sensor with spoofing detection
CN103116744B (zh) * 2013-02-05 2016-04-13 浙江工业大学 基于mrf和svm-knn分类的假指纹检测方法
CA2900479A1 (en) 2013-02-06 2014-08-14 Sonavation, Inc. Biometric sensing device for three dimensional imaging of subcutaneous structures embedded within finger tissue
US10430631B2 (en) * 2013-07-16 2019-10-01 The Regents Of The University Of California MUT fingerprint ID system
KR102402146B1 (ko) * 2015-04-21 2022-05-26 삼성전자주식회사 지문 감지 방법 및 장치

Patent Citations (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5456256A (en) 1993-11-04 1995-10-10 Ultra-Scan Corporation High resolution ultrasonic imaging apparatus and method
WO2001024700A1 (en) 1999-10-07 2001-04-12 Veridicom, Inc. Spoof detection for biometric sensing systems
US20030202687A1 (en) 2002-04-29 2003-10-30 Laurence Hamid Method for preventing false acceptance of latent fingerprint images
US20070030475A1 (en) 2003-04-04 2007-02-08 Lumidigm, Inc. White-light spectral biometric sensors
US20050259850A1 (en) 2003-08-26 2005-11-24 Toshishige Shimamura Surface shape recognition sensor
US20050084154A1 (en) * 2003-10-20 2005-04-21 Mingjing Li Integrated solution to digital image similarity searching
WO2006042144A2 (en) 2004-10-07 2006-04-20 Ultra-Scan Corporation Ultrasonic fingerprint scanning utilizing a plane wave
US20100251824A1 (en) * 2004-10-07 2010-10-07 Schneider John K Ultrasonic Fingerprint Scanning Using a Plane Wave
US7505613B2 (en) 2005-07-12 2009-03-17 Atrua Technologies, Inc. System for and method of securing fingerprint biometric systems against fake-finger spoofing
US20070092115A1 (en) 2005-10-26 2007-04-26 Usher David B Method and system for detecting biometric liveness
US7738681B1 (en) 2005-11-01 2010-06-15 Hewlett-Packard Development Company, L.P. Fingerprint and physical attribute detection
EP1835438A1 (en) 2006-03-13 2007-09-19 Fujitsu Ltd. Fingerprint authenticating system
US20070230754A1 (en) 2006-03-30 2007-10-04 Jain Anil K Level 3 features for fingerprint matching
US7804984B2 (en) 2006-07-31 2010-09-28 Lumidigm, Inc. Spatial-spectral fingerprint spoof detection
US20080253625A1 (en) 2006-10-10 2008-10-16 Stephanie Schuckers Multi-Resolutional Texture Analysis Fingerprint Liveness Systems and Methods
US20090316963A1 (en) 2008-06-19 2009-12-24 Authentec, Inc. Software based method for finger spoof detection and related devices
US20100008552A1 (en) 2008-07-14 2010-01-14 Union Community Co., Ltd. Apparatus for distinguishing living body fingerprint and method thereof
US20100113952A1 (en) 2008-11-03 2010-05-06 Raguin Daniel H Apparatus and method for the identification of fake fingerprints
US20130058544A1 (en) 2009-09-22 2013-03-07 Thales Method for Detecting a Fake Finger for Fingerprint Acquisition Software
US20120263355A1 (en) * 2009-12-22 2012-10-18 Nec Corporation Fake finger determination device
US20110158535A1 (en) * 2009-12-24 2011-06-30 Canon Kabushiki Kaisha Image processing apparatus and image processing method
US20130202182A1 (en) 2011-08-09 2013-08-08 Lumidigm, Inc. Methods and systems for estimating genetic characteristics from biometric measurements
WO2014012467A1 (zh) 2012-07-20 2014-01-23 中国人民解放军军事医学科学院毒物药物研究所 间二芳基苯胺类或吡啶胺类化合物、其制备方法及医药用途
US20140226879A1 (en) 2013-02-08 2014-08-14 Apple Inc. Fingerprint Imaging and Quality Characterization
US20140241596A1 (en) 2013-02-28 2014-08-28 The Regents Of The University Of California Apparatus and Method for Capturing a Vital Vascular Fingerprint
US20160070967A1 (en) 2014-09-05 2016-03-10 Qualcomm Incorporated Multi-Stage Liveness Determination

Non-Patent Citations (14)

* Cited by examiner, † Cited by third party
Title
Arulkumaran T., et al., "Fingerprint Based Age Estimation Using 2D Discrete Wavelet Transforms and Principal Component Analysis", International Journal of Advanced Research in Electrical Electronics and Instrumentation Engineering, Mar. 3, 2013 (Mar. 3, 2013), pp. 1060-1066, XP055227078, Retrieved from the Internet: URL: http://www.ijareeie.com/upload/march/14-Fingerprint%20Based.pdf [retrieved on Nov. 10, 2015].
Bossen A., et al., "Internal Fingerprint Identification With Optical Coherence Tomography", IEEE Photonics Technology Letters, vol. 22, No. 7, Apr. 1, 2010, pp. 507-509.
Drahansky M., "Liveness Detection in Biometrics," Advanced Biometric Technologies, 2011, 21 pages.
Galbally et al., "A high performance fingerprint liveness detection method based on quality related features", 2010 Elsevier, 2 cover pages + pp. 311-321. *
International Preliminary Report on Patentability-PCT/US2015/048583-ISA/EPO-Aug. 30, 2016.
International Prelimminary Report on Patentability-PCT/US2015/048581-ISA/EPO-Dec. 8, 2016.
International Search Report and Written Opinion-PCT/US2015/048581-ISA/EPO-Nov. 23, 2015.
International Search Report and Written Opinion-PCT/US2015/048583-ISA/EPO-Nov. 11, 2015.
Mela G., et al.,"Fingerprint Recognition Using Haar Wavelet Transform and Local Ridge Attributes Only," Jan. 31, 2014 (Jan. 31, 2014), XP055292736, Retrieved from the Internet: URL: http://www.ijarcsse.com/docs/papers/Volume-4/1-January2014/V4I1-0103.pdf [retrieved on Aug. 2, 2016].
Toth B, "Biometric Liveness Detection", Information Security Bulletin, Chi Publishing, Oct. 1, 2005, XP007906384, vol. 10, pp. 291-298.
U.S. Corrected Notice of Allowability dated Mar. 8, 2017, issued in U.S. Appl. No. 14/845,149.
U.S. Notice of Allowance dated Feb. 16, 2017, issued in U.S. Appl. No. 14/845,149.
U.S. Office Action dated Oct. 27, 2016, issued in U.S. Appl. No. 14/845,149.
Warwante et al., Wavelet Based Fingerprint Liveness Detection), International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622, vol. 2, Issue 2,Mar.-Apr. 2012, pp. 1643-1645. *

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10776604B2 (en) 2006-04-26 2020-09-15 Aware, Inc. Fingerprint preview quality and segmentation
US10325137B2 (en) * 2006-04-26 2019-06-18 Aware, Inc. Fingerprint preview quality and segmentation
US11250239B2 (en) 2006-04-26 2022-02-15 Aware, Inc. Fingerprint preview quality and segmentation
US9953233B2 (en) 2014-09-05 2018-04-24 Qualcomm Incorporated Multi-stage liveness determination
US11374929B2 (en) 2017-03-21 2022-06-28 Global E-Dentity, Inc. Biometric authentication for an augmented reality or a virtual reality device
US10721228B2 (en) 2017-03-21 2020-07-21 Global E-Dentity Inc. Biometric authentication of individuals utilizing characteristics of bone and blood vessel structures
US10880303B2 (en) 2017-03-21 2020-12-29 Global E-Dentity, Inc. Real-time COVID-19 outbreak identification with non-invasive, internal imaging for dual biometric authentication and biometric health monitoring
US10547611B2 (en) 2017-03-21 2020-01-28 YouaretheID, LLC Biometric authentication of individuals utilizing characteristics of bone and blood vessel structures
US11277405B2 (en) 2017-03-21 2022-03-15 Global E-Dentity, Inc. Dual biometric authentication and biometric health monitoring using chemosensory and internal imaging data
US11368455B2 (en) 2017-03-21 2022-06-21 Global E-Dentity, Inc. Biometric authentication of individuals utilizing characteristics of bone and blood vessel structures
US10135822B2 (en) 2017-03-21 2018-11-20 YouaretheID, LLC Biometric authentication of individuals utilizing characteristics of bone and blood vessel structures
US11335116B2 (en) 2017-07-20 2022-05-17 Seung Jin Lee Apparatus and method for identifying fake fingerprint by using impedance
US10685204B2 (en) 2018-07-27 2020-06-16 Qualcomm Incorporated Biometric age estimation via ultrasonic imaging
US12028347B2 (en) 2019-09-25 2024-07-02 Amod Ashok Dange System and method for enabling a user to obtain authenticated access to an application using a biometric combination lock
US11582032B2 (en) 2019-09-25 2023-02-14 Amod Ashok Dange System and method for sharing user preferences without having the user reveal their identity
US12026247B2 (en) 2019-09-25 2024-07-02 Amod Ashok Dange System and method for enabling a user to create an account on an application or login into the application without having the user reveal their identity
US12072963B2 (en) 2019-09-25 2024-08-27 Amod Ashok Dange System and method for affixing a signature using biometric authentication
US12079367B2 (en) 2019-09-25 2024-09-03 Amod Ashok Dange System and method for enabling social network users to grant their connections granular visibility into their personal data without granting the same visibility to the network operator
US11468712B2 (en) 2020-01-09 2022-10-11 AuthenX Inc. Liveness detection apparatus, system and method
US12051267B2 (en) 2020-04-29 2024-07-30 Qualcomm Incorporated Fingerprint sensor with detection of latent fingerprints
US20230101088A1 (en) * 2021-09-28 2023-03-30 Samsung Display Co., Ltd. Method and apparatus for restoring image of fingerprint sensor
US12217396B2 (en) * 2021-09-28 2025-02-04 Samsung Display Co., Ltd. Method and apparatus for restoring image of fingerprint sensor
US20230360653A1 (en) * 2022-05-04 2023-11-09 Board Of Trustees Of Michigan State University Speaker Verification System Using Ultrasound Energy In Human Speech
US12347440B2 (en) * 2022-05-04 2025-07-01 Board Of Trustees Of Michigan State University Speaker verification system using ultrasound energy in human speech

Also Published As

Publication number Publication date
WO2016037078A1 (en) 2016-03-10
WO2016037077A1 (en) 2016-03-10
US20160070967A1 (en) 2016-03-10
JP2017535303A (ja) 2017-11-30
US9639765B2 (en) 2017-05-02
KR102636547B1 (ko) 2024-02-13
KR20220154855A (ko) 2022-11-22
CN106663198A (zh) 2017-05-10
BR112017004379B1 (pt) 2023-01-31
CN106796650A (zh) 2017-05-31
CA2957218A1 (en) 2016-03-10
KR102467916B1 (ko) 2022-11-15
EP3189471B1 (en) 2018-08-01
CA2957218C (en) 2023-02-14
US9953233B2 (en) 2018-04-24
BR112017004379A2 (pt) 2017-12-05
JP6656228B2 (ja) 2020-03-04
KR20170048390A (ko) 2017-05-08
AU2015311785B2 (en) 2020-05-21
CN106796650B (zh) 2020-09-22
US20170200054A1 (en) 2017-07-13
US20160070968A1 (en) 2016-03-10
AU2015311785B9 (en) 2020-05-28
EP3189471A1 (en) 2017-07-12
AU2015311785A1 (en) 2017-03-02
EP3189470A1 (en) 2017-07-12

Similar Documents

Publication Publication Date Title
US9633269B2 (en) Image-based liveness detection for ultrasonic fingerprints
Wang et al. Design and implementation of Log-Gabor filter in fingerprint image enhancement
Yang et al. A fingerprint verification algorithm using tessellated invariant moment features
EP2883190B1 (en) Texture features for biometric authentication
CN104700018B (zh) 一种用于智能机器人的识别方法
Raghavendra et al. Presentation attack detection algorithms for finger vein biometrics: A comprehensive study
WO2018023884A1 (zh) 一种身份识别装置及方法
WO2009105324A2 (en) Sketch-based password authentication
CN106709450A (zh) 一种指纹图像识别方法及系统
Zanganeh et al. Partial fingerprint matching through region-based similarity
WO2018082011A1 (zh) 活体指纹识别方法及装置
Krishna Prasad et al. fingerprint image segmentation: A review of state of the art techniques
Hilles et al. Latent fingerprint enhancement and segmentation technique based on hybrid edge adaptive DTV model
Naser et al. Artificial Neural Network-Based Fingerprint Classification and Recognition.
CN105138886B (zh) 机器人生物体征识别系统
Hany et al. Speeded-Up Robust Feature extraction and matching for fingerprint recognition
Raja et al. Prognostic evaluation of multimodal biometric traits recognition based human face, finger print and iris images using ensembled SVM classifier
Panetta et al. LQM: Localized quality measure for fingerprint image enhancement
Rahulkar et al. Fast discrete curvelet transform based anisotropic iris coding and recognition using k-out-of-n: A fused post-classifier
Yuan et al. Fingerprint liveness detection adapted to different fingerprint sensors based on multiscale wavelet transform and rotation-invarient local binary pattern
Silpamol et al. Detection and rectification of distorted fingerprints
Sharma et al. Experimental analysis of a multimodal biometric system using preprocessing and feature extraction techniques and their impact on analytical results
Akintoye et al. Challenges of finger vein recognition system: a theoretical perspective
Babatunde et al. A multi-level model for fingerprint image enhancement
Ogundepo et al. Development of a real time fingerprint authentication/identification system for students’ record

Legal Events

Date Code Title Description
AS Assignment

Owner name: QUALCOMM INCORPORATED, CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GU, JIN;CHEN, MING YU;DU, ELIZA YINGZI;AND OTHERS;SIGNING DATES FROM 20151030 TO 20160122;REEL/FRAME:037603/0288

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

STCF Information on status: patent grant

Free format text: PATENTED CASE

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 8